Fri, Mar 29, 4:13 AM CDT

Welcome to the Poser 12 Forum

Forum Moderators: nerd, donnena, CHMedia Forum Coordinators: nerd

Poser 12 F.A.Q (Last Updated: 2024 Feb 28 7:20 pm)



Welcome to the Poser Forums! Need help with these versions, advice on upgrading? Etc...you've arrived at the right place!


Looking for Poser Tutorials? Find those HERE



Subject: Spotlight falloff bug?


AcePyx ( ) posted Thu, 11 February 2021 at 2:08 AM · edited Thu, 28 March 2024 at 1:42 AM

Created a spotlight to address distant light shadow blur bug. Place spot at z10 y10 intensity 100 scale 100 - render Move spot to z1000 y1000 intensity 100 scale 10000

Light actually INCREASES intensity as it scales up and moves away. I would expect an increase in shadows as the scale decreases and the light moves closer, but surely the falloff should be significant at that distance, and the intensity should decrease with scale?

Distance 10 light close.png

Distance 1000 light far.png


bagginsbill ( ) posted Thu, 11 February 2021 at 5:51 AM · edited Thu, 11 February 2021 at 5:56 AM

You're asking about illuminance (how intense is the light arriving at a surface) as a function of distance AND size (scale) of light source, keeping all other things equal.

The lighting equation has many terms but we'll use a simplified one where the variables are roughly just the ones you were changing.

Letting I stand for illuminance, B stand for the brightness of the light source, R is the distance to a light source, and S is the size (scale) of that light source we have

I = B (S^2) / (R^2)

In English, the illuminance is proportional to the brightness of the light, the square of the scale, and inversely proportional to the square of the distance.

Your first case you claim the distance is 10. But the coordinates you gave (z10 y10) would make the distance roughly 14. As well you didn't mention an x offset but I can clearly see that the light isn't lined up on the Z axis so I'll assume x10 as well. Therefore the distance squared (R^2) is 10^2 + 10^2 + 10^2 or 300.

I = B (100^2) / 300 = 33.3 B

Your second case we have z1000 y1000 and I'll assume x is still 10 (looks that way based on the pattern of specular highlights and shadows). This gives a distance squared (R^2) of 1000^2 + 1000^2 + 10^2 or 2000100. You also said the scale is now 10000.

I = B (10000^2) / 2000100 = 50 B

50 is bigger than 33.3, therefore we expect the figure to be brighter.

However, I didn't verify my assumptions about the lighting terms in the version of Poser you're running. There COULD be a bug or other factor I didn't take into account.

For example, I'm not absolutely sure the brightness of the light source is unaffected by changing its scale. If it is, then I'm missing some of the correct equation.


Renderosity forum reply notifications are wonky. If I read a follow-up in a thread, but I don't myself reply, then notifications no longer happen AT ALL on that thread. So if I seem to be ignoring a question, that's why. (Updated September 23, 2019)


Y-Phil ( ) posted Thu, 11 February 2021 at 8:02 AM

bagginsbill posted at 8:01AM Thu, 11 February 2021 - #4412520

You're asking about illuminance (how intense is the light arriving at a surface) as a function of distance AND size (scale) of light source, keeping all other things equal.

The lighting equation has many terms but we'll use a simplified one where the variables are roughly just the ones you were changing.

Letting I stand for illuminance, B stand for the brightness of the light source, R is the distance to a light source, and S is the size (scale) of that light source we have

I = B (S^2) / (R^2)

In English, the illuminance is proportional to the brightness of the light, the square of the scale, and inversely proportional to the square of the distance.

Your first case you claim the distance is 10. But the coordinates you gave (z10 y10) would make the distance roughly 14. As well you didn't mention an x offset but I can clearly see that the light isn't lined up on the Z axis so I'll assume x10 as well. Therefore the distance squared (R^2) is 10^2 + 10^2 + 10^2 or 300.

I = B (100^2) / 300 = 33.3 B

Your second case we have z1000 y1000 and I'll assume x is still 10 (looks that way based on the pattern of specular highlights and shadows). This gives a distance squared (R^2) of 1000^2 + 1000^2 + 10^2 or 2000100. You also said the scale is now 10000.

I = B (10000^2) / 2000100 = 50 B

50 is bigger than 33.3, therefore we expect the figure to be brighter.

However, I didn't verify my assumptions about the lighting terms in the version of Poser you're running. There COULD be a bug or other factor I didn't take into account.

For example, I'm not absolutely sure the brightness of the light source is unaffected by changing its scale. If it is, then I'm missing some of the correct equation.

Thank you so much bagginsbill: I was truly missing your excellent explanations... <3

PhYl.


Win10 on i7 8700K@4.3Ghz, 64Gb, Asus TUF Gaming RTX 4070 OC Edition, 2x 2Tb ssd + 6+4Tb hd  + 1x 8Tb hd + 1 10T NAS, Poser 11, Poser 12  and now Poser 13 


ChromeStar ( ) posted Thu, 11 February 2021 at 11:07 AM

Scale definitely affects the amount of light for area lights, not sure about spots.


ghostship2 ( ) posted Thu, 11 February 2021 at 12:24 PM

@AcePyx you never tell us what attenuation you are using for your spot light. That would affect this greatly.

W10, Ryzen 5 1600x, 16Gb,RTX2060Super+GTX980, PP11, 11.3.740


AcePyx ( ) posted Sat, 13 February 2021 at 7:59 PM

bagginsbill posted at 7:57PM Sat, 13 February 2021 - #4412520

You're asking about illuminance (how intense is the light arriving at a surface) as a function of distance AND size (scale) of light source, keeping all other things equal.

The lighting equation has many terms but we'll use a simplified one where the variables are roughly just the ones you were changing.

Letting I stand for illuminance, B stand for the brightness of the light source, R is the distance to a light source, and S is the size (scale) of that light source we have

I = B (S^2) / (R^2)

In English, the illuminance is proportional to the brightness of the light, the square of the scale, and inversely proportional to the square of the distance.

Your first case you claim the distance is 10. But the coordinates you gave (z10 y10) would make the distance roughly 14. As well you didn't mention an x offset but I can clearly see that the light isn't lined up on the Z axis so I'll assume x10 as well. Therefore the distance squared (R^2) is 10^2 + 10^2 + 10^2 or 300.

I = B (100^2) / 300 = 33.3 B

Your second case we have z1000 y1000 and I'll assume x is still 10 (looks that way based on the pattern of specular highlights and shadows). This gives a distance squared (R^2) of 1000^2 + 1000^2 + 10^2 or 2000100. You also said the scale is now 10000.

I = B (10000^2) / 2000100 = 50 B

50 is bigger than 33.3, therefore we expect the figure to be brighter.

However, I didn't verify my assumptions about the lighting terms in the version of Poser you're running. There COULD be a bug or other factor I didn't take into account.

For example, I'm not absolutely sure the brightness of the light source is unaffected by changing its scale. If it is, then I'm missing some of the correct equation.

That's an invaluable insight Baggins thank you. I had no idea it worked that way. Now I can adjust the formula to achieve roughly similar values as distance increases. Really appreciate this.


AcePyx ( ) posted Sat, 13 February 2021 at 8:03 PM

ghostship2 posted at 8:02PM Sat, 13 February 2021 - #4412566

@AcePyx you never tell us what attenuation you are using for your spot light. That would affect this greatly.

Hey Ghost, I don't even know what attenuation is, and I certainly haven't changed it from one to another. What does it do?


AcePyx ( ) posted Sat, 13 February 2021 at 8:04 PM

ChromeStar posted at 8:03PM Sat, 13 February 2021 - #4412549

Scale definitely affects the amount of light for area lights, not sure about spots

that's what I thought, and I just naturally thought it was like the sun - the bigger the light, the lower the brightness..


ChromeStar ( ) posted Sun, 14 February 2021 at 1:18 AM

Bigger area lights produce more light. I just tried changing the scale on a spot light, though, and it didn't seem to make any difference at all.

Regarding Attenuation, look in the light properties, below where you select whether it is a spot, point, infinite light, etc. It's the bottom setting. The default is Constant, which means the light is providing the same amount of illumination regardless of distance. Easy to work with, not very realistic in most cases. There is also an inverse linear, in which case it falls off linearly with distance, and inverse square which falls off with the square of distance. If you want the light to fall off with distance, pick one of those two -- inverse square is generally more realistic.


RedPhantom ( ) posted Sun, 14 February 2021 at 7:07 AM
Forum Moderator

larger area lights don't produce more light, they produce softer shadows so it looks brighter, but the non-shadowed colors are the same. The same with spotlights and point lights. If you were to get more light, the colors would start to wash out. That part isn't visible in the samples above because the figure is all white already.


Available on Amazon for the Kindle E-Reader Monster of the North and The Shimmering Mage

Today I break my own personal record for the number of days for being alive.
Check out my store here or my free stuff here
I use Poser 13 and win 10


AcePyx ( ) posted Sun, 14 February 2021 at 12:35 PM

ChromeStar posted at 12:34PM Sun, 14 February 2021 - #4412730

Bigger area lights produce more light. I just tried changing the scale on a spot light, though, and it didn't seem to make any difference at all.

Regarding Attenuation, look in the light properties, below where you select whether it is a spot, point, infinite light, etc. It's the bottom setting. The default is Constant, which means the light is providing the same amount of illumination regardless of distance. Easy to work with, not very realistic in most cases. There is also an inverse linear, in which case it falls off linearly with distance, and inverse square which falls off with the square of distance. If you want the light to fall off with distance, pick one of those two -- inverse square is generally more realistic.

Oh yeah, now I know. So long since I used it, I completely forgot. Thanks!


AcePyx ( ) posted Sun, 14 February 2021 at 12:36 PM

RedPhantom posted at 12:36PM Sun, 14 February 2021 - #4412739

larger area lights don't produce more light, they produce softer shadows so it looks brighter, but the non-shadowed colors are the same. The same with spotlights and point lights. If you were to get more light, the colors would start to wash out. That part isn't visible in the samples above because the figure is all white already.

That's interesting. So I wonder how large I need to make them to have the same shadow effect as an infinite light?


ghostship2 ( ) posted Sun, 14 February 2021 at 6:44 PM

area lights wont ever look like inf lights. If you want a light to illuminate the entire scene and cast realistic looking shadows and highlights then you'll need to use the background as a light source as I've show in other threads. background light 1.jpgbackground again.jpg

W10, Ryzen 5 1600x, 16Gb,RTX2060Super+GTX980, PP11, 11.3.740


ghostship2 ( ) posted Sun, 14 February 2021 at 6:45 PM

the alternative is to use a dome and 1 infinite light for the sun.

W10, Ryzen 5 1600x, 16Gb,RTX2060Super+GTX980, PP11, 11.3.740


AcePyx ( ) posted Tue, 16 February 2021 at 10:55 AM

ghostship2 posted at 10:51AM Tue, 16 February 2021 - #4412779

the alternative is to use a dome and 1 infinite light for the sun.

Thank you for your helpful comments Ghost. Yes, I was using snarlygribbly's skydome, but of course the script doesn't work in Poser 12. :-(

You say that I should use the background as a light source. I'm not sure I'm understanding. HDRIs don't cast shadows do they? Are you talking about a 360 degree background?


hborre ( ) posted Tue, 16 February 2021 at 5:11 PM

That is exactly what ghostship2 is suggesting. A high-quality 360-degree panorama HDR provides a natural realistic lighting condition. That is what Snarly's EZDome provides, using a two-dome setup, one dome providing illumination and the other dome providing a background image. ATM, the image nodes in P12 are broken, they cannot project a background nor scene illumination. Until that is repaired, EXR files will have to do except not everyone have the software to convert HDRI to EXR.


AcePyx ( ) posted Tue, 16 February 2021 at 5:42 PM

hborre posted at 5:41PM Tue, 16 February 2021 - #4412972

That is exactly what ghostship2 is suggesting. A high-quality 360-degree panorama HDR provides a natural realistic lighting condition. That is what Snarly's EZDome provides, using a two-dome setup, one dome providing illumination and the other dome providing a background image. ATM, the image nodes in P12 are broken, they cannot project a background nor scene illumination. Until that is repaired, EXR files will have to do except not everyone have the software to convert HDRI to EXR.

Thank you. What do you mean by "the image nodes are broken?" In what way?


ghostship2 ( ) posted Tue, 16 February 2021 at 5:54 PM

HDRI's got broke in the last public release. I'm using an older beta release because I gotta get stuff done.

W10, Ryzen 5 1600x, 16Gb,RTX2060Super+GTX980, PP11, 11.3.740


ghostship2 ( ) posted Tue, 16 February 2021 at 6:01 PM

@Acepyx if you open up the material tab and look at "Object" you will notice the two on the bottom of the list. "Background" and "Atmosphere." The background can be used to illuminate your scene. You can use any image you want and your figures and props will take on the lighting characteristics of the background image. There will be no bright highlights or dark shadows, just an even lighting that matches your background image. If you use an HDRI that is truly high dynamic range then you will also get bright highligts and dark shadows projected from the HDRI. The HDRI that I used in my example up there includes the sun and that is projecting sunlight on the figure and casting shadows like it should.

W10, Ryzen 5 1600x, 16Gb,RTX2060Super+GTX980, PP11, 11.3.740


ghostship2 ( ) posted Tue, 16 February 2021 at 6:24 PM

here is another example of using an HDRI. First example the sun is to her left and bouncing on her shoulder. Middle, I took the same image and bumped up the exposure with Poser's post FX to about 2.7. The last one I rotated the HDRI so the sunlight is shining directly on her face. I like the middle one. light 1_Original.jpg

W10, Ryzen 5 1600x, 16Gb,RTX2060Super+GTX980, PP11, 11.3.740


AcePyx ( ) posted Tue, 16 February 2021 at 7:36 PM

ghostship2 posted at 7:35PM Tue, 16 February 2021 - #4412981

here is another example of using an HDRI. First example the sun is to her left and bouncing on her shoulder. Middle, I took the same image and bumped up the exposure with Poser's post FX to about 2.7. The last one I rotated the HDRI so the sunlight is shining directly on her face. I like the middle one.

Thank you. This was using snarlygribbly's skydome correct?


ghostship2 ( ) posted Tue, 16 February 2021 at 8:35 PM

no. Domes will not produce "sunlight," only ambient light. I used Poser's "background" that is always there in your scene. click on the materials tab then "object" and select "background." use the material setup that I posted up this thread.

W10, Ryzen 5 1600x, 16Gb,RTX2060Super+GTX980, PP11, 11.3.740


ghostship2 ( ) posted Tue, 16 February 2021 at 8:36 PM

make sure you make the construct invisible so the background can be seen.

W10, Ryzen 5 1600x, 16Gb,RTX2060Super+GTX980, PP11, 11.3.740


ghostship2 ( ) posted Tue, 16 February 2021 at 8:38 PM

jeebus! Did I spell "enhanced" wrong? embarrassing.

W10, Ryzen 5 1600x, 16Gb,RTX2060Super+GTX980, PP11, 11.3.740


AcePyx ( ) posted Wed, 17 February 2021 at 9:32 AM

ghostship2 posted at 9:32AM Wed, 17 February 2021 - #4412994

jeebus! Did I spell "enhanced" wrong? embarrassing.

Lol. Thank sfor the tips!


Privacy Notice

This site uses cookies to deliver the best experience. Our own cookies make user accounts and other features possible. Third-party cookies are used to display relevant ads and to analyze how Renderosity is used. By using our site, you acknowledge that you have read and understood our Terms of Service, including our Cookie Policy and our Privacy Policy.